Learning Reduced Fluid Dynamics

Authors: Zherong Pan, Xifeng Gao, Kui Wu

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Through evaluations on a row of simulation benchmarks, we show that our method reduces the discrepancy by 50-90 percent over conventional reduced models and we outperform PINNs by exactly preserving the time reversibility.
Researcher Affiliation Industry Lightspeed Studios {zrpan,xifgao,kwwu}@global.tecent.com
Pseudocode No The paper mentions outlining an algorithm in Appendix 1 but does not present structured pseudocode or an algorithm block in the main text.
Open Source Code No The paper does not provide any statement or link indicating the public release of source code for the described methodology.
Open Datasets No The paper uses simulation benchmarks like 'Taylor vortices' and 'smoke plume rise' and states, 'Our training dataset for the POD baseline contains N = 8 trajectories using the full-order dynamics (Equation 1)', but does not provide specific access information (link, DOI, repository, or formal citation with authors/year) for these simulation data.
Dataset Splits No The paper mentions training and testing data, but does not explicitly describe a validation dataset split or a cross-validation setup.
Hardware Specification Yes We implement our method using Pytorch with a fluid simulator implemented via native C++ with CPU parallelization, and perform all the computations on an AMD Threadripper 3970X CPU having 32 cores.
Software Dependencies No The paper states 'We implement our method using Pytorch with a fluid simulator implemented via native C++', but does not provide specific version numbers for Pytorch, C++, or any other software dependencies.
Experiment Setup Yes We always use a batch size of 1. The performance of our method is summarized in Table 2. We consider two variants of our method: coupled case, where Ckij is treated as a function C(Uk,Ui,Uj) as discussed in Section , and decoupled case, where Ckij is treated as an antisymmetric independent decision variable. ... We experiment with four parameters ϵ = 0.05, 0.01, 0.001, and 0.0001 and the number of bases is p = 8, 11, 16, and 25, correspondingly. ... we set T = 500, δt = 0.01.